Streaming Task Graph Scheduling for Dataflow Architectures

TL;DR

This paper presents a streaming scheduling approach for dataflow architectures that improves device utilization and speedup by analyzing and partitioning task graphs for efficient execution.

Contribution

It introduces canonical task graphs and a static analysis method to enable effective streaming scheduling on dataflow devices, enhancing performance.

Findings

Streaming scheduling increases speedup over traditional methods.

Device utilization is improved through task graph partitioning.

The approach is validated on synthetic and realistic workloads.

Abstract

Dataflow devices represent an avenue towards saving the control and data movement overhead of Load-Store Architectures. Various dataflow accelerators have been proposed, but how to efficiently schedule applications on such devices remains an open problem. The programmer can explicitly implement both temporal and spatial parallelism, and pipelining across multiple processing elements can be crucial to take advantage of the fast on-chip interconnect, enabling the concurrent execution of different program components. This paper introduces canonical task graphs, a model that enables streaming scheduling of task graphs over dataflow architectures. We show how a task graph can be statically analyzed to understand its steady-state behavior, and we use this information to partition it into temporally multiplexed components of spatially executed tasks. Results on synthetic and realistic workloads show how streaming scheduling can increase speedup and device utilization over a traditional scheduling approach.